Introduction: Technological advances have introduced three-dimensional (3-D) printing as an option for creating tactile maps for people with visual impairments (that is, those who are blind or have low vision), diversifying the types of map products that are available. At the same time, it presents a challenge to map makers to implement designs across multiple production methods. We evaluated map symbols to determine their discriminability across three different materials: microcapsule paper, 3-D printer plastic, and embossed paper. Methods: In a single session lasting less than 90 minutes, participants completed a matching task and provided informal feedback regarding their preferences. We measured speed and accuracy to establish discriminability of map symbols on each of the materials. Eighteen participants were recruited from a referred sample among attendees at the American Council of the Blind annual convention in 2013. Results: Response times were significantly different across the three materials (p < 0.001). Without sacrificing accuracy, response times were faster for the 3-D printed graphics than for either the microcapsule paper (p < 0.001) or the embossed paper (p < 0.001). User preference was divided across the three materials. Some people disliked the "sharp" corners of the 3-D printed symbols, while others preferred their "crisp" edges. Discussion: Our results demonstrate faster discriminability of a set of tactile symbols produced on a 3-D printer compared to those same symbols printed on microcapsule paper, the material for which the symbols were originally designed. Participant feedback reflected preferences both in favor of and against reading symbols produced on the 3-D printer. Implications for practitioners: This article discusses the functional equivalence of tactile symbols produced across multiple production technologies. It addresses two considerations when using 3-D printing to make tactile maps: preparing digital files for printing and the printing work flow. Digital files ready for printing on each of the three materials are available for download (Brittell, Lobben, & Lawrence 2016).